人工免疫算法在排课系统中的应用

发布时间：2018-04-09 01:12

本文选题：多目标优化　切入点：人工免疫　出处：《浙江工业大学》2012年硕士论文

【摘要】：在校务行政上，排课一直都是无法避免的例行工作。随着高校教育事业的发展和招生规模的持续扩大，近年来在校学生的人数也在不断增加，加上学校资源受限的问题，如何进行合理的排课就成为现代教育管理中一项关键的任务。排课作业可以归纳为：如何找出教师、学生、教学设施、规定习惯和上课时间等复杂限制条件下的最优解。它是一个多重约束满足问题的一种复杂的NP-Complete（Non-deterministic Polynomial Time）组合优化问题。由于必须同时考虑许多因素与限制条件，使其成为十分复杂的问题；而利用计算机来解决排课问题，虽已有许多方法被提出，但由于各校系较无一致性的考虑与标准，，目前尚未有可广泛应用于各校系的解决方案。人工免疫系统是一种模仿自然免疫系统功能的智能方法，目前在多目标优化问题中得到了较好的应用效果。本论文针对各种影响排课之因素与限制进行分析，以人工免疫理论为基础，结合了前人的研究成果，将生物免疫进化选择的思想引入到多目标进化算法之中，提出了克隆多目标进化算法，并应用于自动排课系统中。使用Borland公司的Delphi6.0开发了基于克隆多目标进化算法的排课系统，其提供了班级、教师、教室信息的输入、编辑、排课、开课计划文档导入与打印等功能。实验证明，该算法不仅能够较好的解决多目标的约束冲突，而且在处理多目标优化问题上能够得到更高的教师满意度，较传统的几种进化算法均有一定的优势。
[Abstract]:In school administration, class scheduling has always been an unavoidable routine.With the development of university education and the continuous expansion of enrollment scale, the number of students in the school has been increasing in recent years, and the problem of limited resources of the school is also being encountered.How to arrange courses reasonably becomes a key task in modern education management.Homework scheduling can be summed up as: how to find out the optimal solution under complex constraints such as teachers, students, teaching facilities, prescribed habits and class hours.It is a complex NP-Complete(Non-deterministic Polynomial time) combinatorial optimization problem with multiple constraints.Many factors and constraints must be taken into account at the same time, which makes it a very complicated problem. Although many methods have been proposed to solve the problem of arranging courses by computer, there is no uniform consideration and standard among the schools and departments.At present, there is no solution that can be widely used in all departments.Artificial immune system (AIS) is an intelligent method to imitate the function of natural immune system.Based on the artificial immune theory and previous research results, this paper introduces the idea of evolutionary selection of biological immunity into the multi-objective evolutionary algorithm.A clonal multiobjective evolutionary algorithm is proposed and applied to the automatic course scheduling system.A class scheduling system based on clonal multi-objective evolutionary algorithm is developed by using Delphi6.0 of Borland Company. It provides functions of class, teacher, classroom information input, editing, course scheduling, course planning document import and printing.Experimental results show that the proposed algorithm can not only solve the multi-objective constraint conflicts, but also obtain higher teacher satisfaction in dealing with multi-objective optimization problems. This algorithm has some advantages over the traditional evolutionary algorithms.
【学位授予单位】：浙江工业大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：G647;TP301.6

【参考文献】